Great topic, Rahim. Book designers really need to cast their net wider. Despite the hundreds of excellent text faces designed specifically for today's workflow and presses, the old metal stuff, often poorly digitized, is still being used. Here are some of my favorite digital typefaces designed specifically for long-running text:

Rongel
Renard
FF Clifford
Merlo
Iowan Oldstyle
Freight Text
Mercury Text
Chronicle Text
ITC Mendoza
Paperback

And there's no reason to remain married to ye olde oldstyle serifs. Contemporary slabbish and semi-serif designs are just as effective for book text:

FF Tisa
Versa
FF Meta Serif

Thanks for kicking off the list, Stephen.

I saw Versa in a new series of Shakespeare plays published exclusively by Barnes & Noble and the darkness is quite lovely.

What does your darkness-meter say about the following:

FF Maiola
Dolly
http://Gentium Book Basic
Athelas
Musee
http://Dante Pro MT
http://Sabon Next Pro
Feijoa

Rahim

Arno
Arnhem
Chaparral
Cycles
Monticello

Every book face ever released by Font Bureau.

ITC Giovanni. I know some people write Giovanni off as anachronistic or immature in comparison to Slimbach’s other work but it holds up beautifully on cheap paper and reads like a dream under bad lighting.

Whitney Book is an incredible sans for text.
Magma would make a fabulous book face given the right subject.

Are there any contemporary Didones designed with books in mind? I’ve seen Monotype Didot look great in books, but can’t think of any others.

A related thread: Choosing a font for book design?

any contemporary Didones?
Storm’s Walbaum Text.
Prillwitz and FF Acanthus both have an explicit Book/Text weight. ITC Bodoni and H&FJ Didot offer optical sizes, too. Furthermore, maybe Andrade or Gianotten. Idiosyncratic: Antares.

This is old news.
Digital type designers have been on the job since day one (Think Quadraat, Charter, Scala, etc.)
In fact it is one of the central issues of contemporary type design.
Many of my faces address it.
And yet lots of books are still set in wishy-washy versions of Perpetua and Bembo.
Go figure.

Yes, Nick's post reminds me that we should add his excellent revision of Beufort to the list. Do see the PDF at his site.

Here are links to: FF Quadraat, FF Scala, Scala's follow up FF Nexus, and ITC Charter.

Speaking of Bembo, one of Monotype's better efforts in recent years was its careful reexamination of the face as Bembo Book.

Two more: Eason
and Vendetta.

Lexicon
http://www.teff.nl/fonts/lexicon/lexicon.html
Check out the different weights A, B, C, …

Documenta
http://www.dutchtypelibrary.nl/Documenta_rdrct.html

Paperback
http://www.houseind.com/index.php?page=search

Brioni Text
http://www.typotheque.com/fonts/brioni_text

May I put in a word for Bitsream News 706? Perhaps not an ideal bookface in terms of all the bits you need, but very nice and dark.
http://www.fontshop.com/fonts/downloads/bitstream/news_706/

I like the colour of ArnoPro.

I second Merlo!!! Excellante!

& speaking of Bembo Book: what does the designation "VP" mean on the FontShop page Stephen linked to?

As in: "Bembo Book Pro VP"? It is probably something that should be terribly obvious to me; but it ain't right now.

& why did Monotype leave that hideous long-tail R in the basic character set? Don't tyhey look at text composition and see how bad it looks in a line and on a page?

powers

Arno is nice and dark. The OpenType features that are turned on for the italics just slay me in InDesign. I've never seen a typeface dance like that!

I just printed out a pdf sample of ITC Charter: wow! Wasn't expecting it to look so nice.

I really wish Merlo (love the name!) came in OpenType.

Rahim

What does everyone think about the darkness levels of FF Maiola, Athelas, Musee, Dante MT, Sabon Next, and Feijoa?

@Will:

I assume the VP is for Value Pack.

Both Dante and Bulmer came through the fires of conversion to PostScript at Monotype with good color intact. It'd be interesting to know how that happened, why those two came out OK when so many did not. I do recall reading about some work that went into Bulmer.

It has amused me that Bulmer came out OK when it was made for PostScript, for as a hot-metal face we had problems getting it to cast right in all the Monotype shops I worked at. It may even be a better face now than it was in metal. That rarely has happened.

Bulmer, by the way, deserves looking at for book work. It has a lot of components, and it is very nicely compact. I always look at Bulmer when I need to fit a lot of text into a book.

Thanks, Florian.

powers

Thanks so much everyone for the information on this thread. Just marked it as a fave as it's such a useful reference. I'm starting to give serious thought to moving into book design and publishing.

Rahim—yes Charter is superb. Matt Carter has his priorities right.

Will—thanks, that's great news about the Monotype Bulmer. It's my kind of font and I dearly want to use it for books.

j a m e s

I'm constantly seeing pt.12 Garamond. It annoys me to no end.

An opentype Merlo is on the way!

When we were asked to update the classic Swedish typeface Berling by Karl-Erik Forsberg we made thorough investigations and found boxes of unused Berling type in metal at a printing museum in the south of Sweden. From these we made trial prints in 8 pt and it's from these the new Berling Nova was based. It turned out much darker and not as edgy as previous digital (and photo setting) versions (we saved that for the display version). I'm rather happy with the result.

ƒ

Does anyone have any PDFs of FF Meta Serif used in a book (not magazine) that we could print out to see its darkness on a page?

Or does anyone know of a book that is already using it that I could go see at a US bookstore?

Rahim

I'm reading a book set in Adobe Jenson ("Ordinary Wolves" by Seth Kantner, Milkweed Editions, 2004). I'm pleased with the dark color on the page. The book is set (I'm guessing) in 10.5 or maybe 11. Leaded to 16, it might be more open than it needs to be.

In a dim November dawn, on a transit bus, there was enough color for good reading.

I'll have to take a look at the Berling Nova color.

powers

Rahim - Contact FontShop and they will be happy to create a PDF sample for you according to your specs.

Although things have improved* I think we can go darker still. Even Smeijers's ideal (darker than most of the examples above) might still be a bit conservative. In any case, it can't be "old news" if "lots of books are still set in wishy-washy versions", so we've got work to do!

* And not just because people are making revivals smarter.

hhp

Hrant,

I totally agree. Even though the problem has been identified by some and solutions have been offered, the importance of avoiding that problem and actually using those new solutions has yet to penetrate the culture of book typesetting so that it becomes more of an expected norm and less of a surprising exception.

Still, I think it would help to speak out more about the importance of doing this in various magazines, industry publications, conferences, blogs, etc. Consistent communication is key.

Rahim

Stephen,

I just printed the PDF sample Fontshop kindly sent me. The color on FF Meta Serif Book looks awesome! Very contemporary looking, love the x-height.

Rahim

Regarding Clifford:

At FontFeed, FontShop has posted Akira Kobayashi's long essay about how he came to make Clifford and about how he made the face. Go here:

http://fontfeed.com/archives/akira-kobayashi-on-ff-clifford/

About halfway down, the book page that shows a Victorian-era woman is from "A Gift of Light," a study of 19th-c photography (I designed that book for the University of Notre Dame Press in 2002).

It was printed on a dull coated sheet, and Clifford really works well. It has great body weight for text, captions, and backmatter. It is very easy to read.

The book may be hard to find. Libraries with extensive library holdings in photography may have it. But I have heard that it did not sell well, and that the publisher pulped the remaining copies (without asking if I wanted to buy some!)

Clifford and Miller are pretty much my "house faces" these days.

powers

[tracking]

Szabolcs

is miller in opentype?

Yes, but you might want to ask FontBureau what features it has programmed into it. For sure the ligatures will substitute but I would not expect contextual spacing or anything like that. OpenType at it very basic.

Miller is more newsy—Clifford more literary. Miller is just so gorgeous tho. Heck, we need both right?

Mikey :-)

All of the Font Bureau Retail fonts are now available in OpenType format. *However,* note that these are what FB is calling "simple" OpenType fonts -- which basically means that they're just a straightforward format conversion, with minimal OTL features, kinda like Adobe's "Std" OT fonts.

FB's sOT fonts provide the basic OT advantages of being single, cross-platform files with Unicode encoding (although the hacked ligatures, et al., are also still in their original positions for backward compatibility, and then duplicated with proper OT names).

In addition, there are a few basic OTL features, the most important being {liga} which allows all the double-f ligatures to substitute automatically now. The three basic fractions have been written into a {frac} substitution; the three basic superiors are in a {sups} substitution; and the two basic ordinals are in a {ordn} substitution.

But, for instance, no attempt has been made yet to integrate SC fonts with their Romans. That will have to wait for the development of ProOT fonts. AFAIK, the FB ProOT standard is still being finalized. I would imagine that Miller would be high on the list of FB fonts for re-development, but I have no specific knowledge.

BTW, simple OT fonts are the same price point as the other formats. And I believe that if you already have a license for FB fonts in PS or TT format, you can get the equivalent sOT by request. But don't quote me on that. You'd have to contact the FB office directly.

-- Kent.

Thank You Kent!

Glad to be of assistance, Mikey.

P.S. I neglected to mention that in the case of the Miller fonts, specifically, there is also a stylistic alternate {salt} feature to substitute the alternate R and ampersand that Matthew included in the designs.

-- K.

cool, thanks kent for the info!

Finally took a good look at Iowan Oldstyle as mentioned by Stephen way up above (it was on my WordPerfect CD of all places!). It's nice and dark and the shape of the dot over the i's looks like a lemon. Nice touch!

Hey Mikey, what's the verdict on an OpenType Merlo? Any timeline?

Thanks!
Rahim

Do note that Iowan has a hefty x-height,
so it's not suitable for "book" setting.

hhp

Bigger than, say, Karmina?

from Karmina's description:

"Karmina is a text typeface developed mainly for pocket books and budget editions. It was built to withstand the worst printing conditions: low quality papers, high printing speed with web presses and variations in the ink level of the printing press. More…

Some of Karmina’s most representative features are the rather large serifs, intended to work perfectly in small reproduction sizes, the sharpness of the shapes, including some calligraphic reminiscences, and the large and yet graceful ink traps in the acute connections. Structurally, Karmina combines a significantly large x-height with relatively compressed letterforms. The result of these features grants Karmina outstanding legibility and economy."

Do you think Iowan Oldstyle's large x-height and dark color give it similar advantages when used in small sizes and bad printing conditions, leaving aside that it is not economically compact as Karmina?

Just trying to learn,
Rahim

Hrant:

This statement of yours above:

"Do note that Iowan has a hefty x-height,
so it’s not suitable for “book” setting."

is pretty broad. Would you say a bit more?

In general I'd be inclined to agree with you, but my own experience in reading and also in designing a book set in Iowan make me realize that such a sweeping statement needs some backing up.

I'm gonna take a look at Karmina.

powers

> Do note that Iowan has a hefty x-height, so it’s not suitable for “book” setting.

I disagree with that conclusion. I've actually used it in a couple books. One needs to take care with the specs, but it can work fine. It's large x-height and proportions give it a particularly approachable feeling that is not appropriate for all settings, but works well in certain others. In this way, I consider it in a class with Stone Serif.

A friend of mine authored a book that was published in Stone Serif, and I do find the short extenders a problem in that particular layout. It is set 10 (I think) over 13 points at a measure of 26 picas.

Karmina is quite charming, and seems very well-crafted.
And its narrowness makes its large x-height effectively smaller.

> This statement of yours above:
> “Do note that Iowan has a hefty x-height,
> so it’s not suitable for “book” setting.”
> is pretty broad. Would you say a bit more?

Maybe you're right, but my quotes around "book" above hopefully save me from having to retract. :-) I was referring to sizes ideal for reading a long book, as opposed to a publication where other factors (such as economy) take higher precedence. And it's for this reason that "Book" cuts of fonts typically have a smaller x-height* (and lighter color, and tighter spacing). For smaller text (and books that for whatever reason need to be set with small text) Iowan is just fine.

* This is because, as you might agree, extenders
need to be leveraged more in immersive reading.

> I do find the short extenders a problem

Indeed.
Just try to read one of the unfortunately
numerous books set in ITC Garamond.

--
All this leads to a question I've asked myself (and others, actually :-) concerning the relationship of vertical proportions and authenticity: don't you guys think that a large x-height is anathema to certain styles, like Venetian and Garamond? It's interesting to note here that Mandel believed that a font can't really be French unless it has a small x-height.

hhp

>Just try to read one of the unfortunately
numerous books set in ITC Garamond.

As luck would have it, I have been reading essays in a book set in ITC Garamond. It is 9 or 10 over 12 points, set at a measure of 27 picas.

It is in a different category than Stone Sans. The Stone Sans I find a little less comfortable than I would like, but still highly readable. Reading the essays in ITC Garamond in that setting is really a chore. I really think my eyes have a harder time holding the line.

I am old, but still...

It seems to be a characteristic of "book" faces that they are quite "busy", with a lot of detail, such as serifs and terminals that vary from letter to letter. Faces that are more uniform in finish are not so bookish.

Why is this redundancy a requirement for sustained reading?

To Nick's question:

There are people in this discussion who have studied far more thoroughly than I the sciences of making typefaces and of reading them. So I'm going to go out a limb and see what you all make of a few ideas; there'll probably be some words that are "non-tech," but I'll try to make the point.

One: Perhaps the "busy" and the redundancy Nick spoke of are essential to what so many of us want in book faces because they break up the uniformity of letter shapes and thus help them form into specific units (words) more easily. Maybe they also impart some basically recognizable features to specific letters. Some "g-ness" or "q-ness." & those nesses help readers put them together into words.

Two: Perhaps there is some cultural familiarity at work here also. Perhaps readers come to "know" book faces (which are faces with these redundancies and busy parts) from early days of reading, and so these become established as "book faces."

That's as far as I am able to go now. I'm already feeling over my head. But behind it I'm sure are some of the factors that have made me accept some faces and reject others for the work I do. I'm sure–I hope–there is something more to it than "I don't like this type."

To Hrant and my original question to him: Perhaps in my general inclination to agree with you I am just echoing what I was told as a kid compositor. But there's some basis for it. Some years ago I learned that the Monotype house I used had TNR with long descenders. So we designed a quarterly newsletter with those descenders, and the thing did look notably better (letterpress). At this remove I cannot say if it read any better.

I cited a few weeks ago in another thread the 900 pages I recently read in ITC Garamond. It was easy as pie, easy on my eyes. The most recent tome was 700 pages in Adobe Caslon, which I found more difficult to read. What gives there?

Back to work for me.

powers

Well, if it's a requirement, it's not a redundancy! :-)
Variety => Contrast => Information

That said, most "book" faces do have a lot
of mannered, non-functional decoration.

> What gives there?

You should turn yourself in at once! ;-)

Question: Was the topic of the ITC Garamond book
particularly interesting, versus the Caslon one?

hhp

>Why is this redundancy a requirement for sustained reading?

That's a really good question. My take on it is that different aspects have different advantages.

Serifs have four advantages:
1. Help the eye hold the line.
2. Make the ends of strokes easier to detect. Tests have shown that the extremes in a glyph are more salient for reading.
3. I think serifs also help to get even color while having wide enough spacing to relieve crowding.
4. They also probably contribute to a distinctive "word image" better than sans. For example, a "b" and "d" are more different in a serif than a sans.

Some variations in design help evenness of color--for example, making the end of the r heavier than the end of the s.

Some variations help break up the "picket fence"--the tendency of too much vertical emphasis and a dazzle, which is always threatening in roman type, and is bad for text.

I was thinking more along the lines of perceptual response to repetitive stimuli.
So it's like you can actively vary what is processed as signal and what is ignored as noise.

Hrant wonders:

"Question: Was the topic of the ITC Garamond book
particularly interesting, versus the Caslon one?"

Both books were equally fascinating, and moving at times. Garamond was the big McCullough bio of Harry Truman. Calson was the same writer's bio of John Adams. Both verged a bit too much toward hagiography, but that's always a danger of biography.

Caslon types have never pleased me very much. They all seem a bit spiky and sparkly, lacking in the dark color that started this here discussion. I rarely specify a book to be set in any form of Caslon. I know that a participant here is at work on a Caslon; I shall attempt to bring an open mind when I can see it.

I know I'm not "supposed" to like ITC Garamond. Being the serious typographer (ahem) that I am. But it posed no reading problems. Even the few instances of faked small caps did not pester me. Nor the lining figs.

powers

>an open mind when I can see it.

Good!

>So it’s like you can actively vary what is processed as signal and what is ignored as noise.

Well, I don't think that's it. I think that the variation helps the signal--the letter and word shapes--come through easily and strongly. I don't think it's noise.

Speaking of dark book faces with a big x-height, I think you fellas might dig tomorrow's FontFont release. Sneak preview:

OMG! Are you kidding me, Stephen? I discovered and fell in love with FF Milo last year. And now a SERIF?!?!

Can you please share a PDF specimen so we can print it out and feast on it ourselves?

Rahim

The family is now live: FF Milo Serif. The official FontFont PDFs aren't working on the site at the moment. Should by tomorrow morning. In the meantime, here's the sample PDF I made because I just couldn't wait. The text is set in Medium because the Text weight is a little too light for me. Even the Bold might do well for text.

Wow! Thanks so much, Stephen. I really appreciate the PDF. The typeface and the color it produces on the page are lovely, nice and dark like FF Meta Serif or FF Clifford.

(Request: can FontShop provide a PDF specimen just like this for all its fonts? The official one with the characters and a few sentences doesn't provide a good sense of what that font might look like in its native setting.)

Interesting that you said that about the text and medium weights. I said the same thing to myself when I saw the book and medium weights for Galaxie Copernicus. The medium weight for that one looks properly dark in print.

Feijoa also only comes in a medium weight, but any smaller weight would cause a loss in the darkness level.

By the way, is there a name for these typefaces like FF Milo Serif that have slightly larger x-height and low ascenders? I know Copernicus is a 21st century Plantin. What family of resemblance would FF Milo Serif belong to?

Rahim

A very handsome face! Thanks for bringing it up.

But is it fair to call it dark? To me even the Medium doesn't qualify for that descriptor... In which case I have to think the designer doesn't intend it as a dark face - especially accounting for the hefty x-height.

On the other hand, on a low-res printer or
in letterpress it would show more weight.

> is there a name for these typefaces

Goût Hollandois? That "a" is very Dutch too.

hhp

Regarding Iowan> I happen to know first hand Iowan would make fabulous book face for extended reading. It fills this role now. Iowan’s namesake, The Iowa Review , has long passages of eye-roll-inducing prose... thank you Borders poetry periodicals. Nevertheless, If I choose to read such blather it would be easy on the eyes.

Regarding Merlo> As far as I know, Felliciano is still working on opentype versions of his library. I would shoot him a line of encouragement to let him know that Merlo is very much on everyone’s minds. I also suggest examining and printing this specimen from Vllg so you can see how dark and lovely the font really is: http://vllg.com/PDFs/vllg_Merlo.pdf

Milo Serif is another gem from Michael Abbink. I wonder if he is still working on Router, an in-progress face mentioned on Daidala?:

http://www.daidala.com/router.html

Mikey :-)

Shall we add Nikola Djurek's Tempera Biblio to the list of dark typefaces?

Rahim

In recent years, Robert Slimbach has been more concerned about ensuring the text weights are heavy enough. Actually, it's always been an important topic, but I think fashion has drifted in that direction lately.

Unfortunately, we also don't proof with an imagesetter as much as we used to. When I got to Adobe, the type group had its own imagesetter down the hall, and we would crank out pages and pages of proofs to ensure that the weights of the text sizes were reasonable. Some time ago, we moved from one floor to another and lost the old imagesetter, and Adobe didn't want to keep chugging through the chemicals used for processing, so now we only get to proof that way once or twice during development. As everyone knows, laser printer proofs require some mental adjustments (but if anyone is good at understanding that, it's Robert).

It was nice in the MM era, when one could just tweak the weight axis and have whatever they liked. Of course, now, we still develop MMs and generate instances, but the end user doesn't get to choose. Recent Slimbach faces have offered good options though, I think. The medium weight of Garamond Premier was created specifically to offer something for more substantial text settings. (The text in the Garamond Premier Pro specimen book is set in the medium weight.)

Brioni, Marlene, Amalia from http://www.typonine.com

Reviving this old thread, with a new(?) question:

Is it generally a bad idea to use a darker-than-regular weight of face for book use?

I ask because I'm selecting a house face for a small startup publishing imprint, and I keep coming back to this thread, as I'm looking for something on the dark side. One option I'm drawn to is Garamond Premier Pro Medium – the regular weight seems too light. But maybe the medium would look too heavy in such an extended setting? It's hard for me to tell without seeing an extended specimen. Perhaps someone here has experience in this area.

While I'm commenting, I'll agree with Stephen that Bembo Book looks really nice. I'm also considering Feliciano's Merlo and Rongel, which have just the kind of darkness I want, but the open-bowl P's are holding me back so far...

Bit late to the party.

I also agree book typefaces should be more black and clearer with enough character to hold you reading for without getting bored.

My font ARS Descendiaan was designed specifically with that in mind or book paper.

and some examples:
http://arstype.tumblr.com/tagged/ARS_Descendiaan

Zach, when we talk about a darker face for book, it's more in it's contrast and proportions. A medium weight like the Garamond you mention can work very well for books, since it has enough contrast to hold well on rougher paper.
However, most of the times at the book publishers we chose not to use the Adobe Garamond cause it was simply to "beautiful" for book text and too perfectly cut that somehow made it more of a display face or for larger sizes.

The best would be to print some text with several fonts you want to check out on the actual type of paper you plan to print your books on, many book printers have already made samples, or can do it for you if you plan a long relationship with them. That would be the most ideal test.

About the Bembo, ALL the Bembo's are too light!! (well almost all except some rare cuts you won't be able to find anywhere). I don't understand why they still keep using it for books. I was once asked to design a 6pt. Bembo to meet with the Publisher's demand for small sizes, to match reproductions of Older publications.

One thing I learned is the the Publisher's world is quite conservative and many have their own specific punctuation rules and demands from a book-typeface, that it's really hard to get in there - any typeface that doesn't fit those rules get's ignored. It almost demands custom typefaces for each publisher.. phew :)

Now that you've revived this topic: I designed Williams Caslon Text to get a "dark but open" look for text, you could add this to the list of fonts to consider for those who find many text types too light or weak in color.

I’m glad this topic has been running for so long (it started about a year and a half ago!). I agree with Nick Shinn’s comment from back near the beginning: you can’t force people to do good design. Some good digital book faces have been around for about 15 years, but most designers don’t realize that it’s important to use (and pay for!) them. Some of my students recently produced a newspaper using ITC Fenice for all the text (without asking me first). It’s dreadful, spidery and impossible to read more than a few sentences without getting a headache.
I remember a conference in (I think) New York City at the end of the 1980s. Monotype put it on and there was discussion involving me, David Berlow, and a few others about optical sizes for digital type. We were just starting to get small caps and oldstyle figures in PostScript type back then. I had experimented with adding stroke weight (dipping in “digital chocolate”) with a few faces, with varying (mostly bad) results. I'm so glad that type is now better than it has ever been since Gutenberg.

Indeed. Technology and quality of typefaces have improved immensely, certainly since Gutenberg. The possibilities now are almost endless when it comes to producing good typefaces to a variety of needs and demands. I even like the still existing practice of designing optical sizes and the quality has improved in that field as well.
You could argue about the actual *need* for all this improvement when you think about the world getting more and more digital and the *value* for the craftsmanship and the amount of work put into certain typefaces getting lower by the year.
Not to mention that print quality has improved quite a bit since then, and it's really more the extreme ends of point sizes that need addressing when it comes to optical sizes.

All this must effect the type of education students get in schools when it comes to typography. Plus, students are not entirely clueless, they see what's going on around them, the vast majority of things to look up to and inspire by are digital or in some digital form or other.. so how do we suppose they'll be interested in this rather old-fashioned practice in legibility and optical sizes and even good typography. Let alone appreciate the (price) value of it.

There are several magazines and publications out there that seem to benefit from this available practice in type-design and use it quite well in most parts (look at the recent redesign of Interview, or T. Thy both have an art-direction that is typography-aware. But is it really from "Need" or "Necessity", or rather the Classical or Sentimental quality they are after.

I hate to use the word Novelty in this context, but I can't help but thinking this is more what's on the minds of the designer/art director these days, or the aspiring designers/art directors out there.

There are new and harder challenges ahead (and present) when it comes to typography and technology, and considering the available "solutions", although some being quite exciting, we still have a long way to go.

And go put a price tag on that. Yikes!

What about the concept of Optical sizes and legibility for Screen (vs Print)?

Yes William, I actually ended up buying your Caslon for this project – it feels darker than Adobe Caslon, and I felt like a Caslon was called for. (This is to set 19th-century novels, which I'm republishing with a twist.)

Glad it worked for you!

Why would you use an 18th century face for 19th century novels?
That seems more like an inadvertent anachronism, rather than a deliberate twist.

Because the target audience is not the historically
pedantic subset in the micro-niche of type designers?

hhp

Nick, Williams Caslon is clearly 21st century, not 18th, though taking inspiration from the 18th. But there were lots of books set in "old style" fonts in the later 19th century (adaptations of Caslon) and toward the end in actual Caslon revivals. There seems to have been a lot of publications in both Scotch and Old Styles.

In any case, I don't think that as a rule a strict "historicism" in setting books is a good idea.

Yeah, poor Bringhurst by himself cannot support the entire book industry...

hhp

Because the target audience is not the historically
pedantic subset in the micro-niche of type designers?

Exactly. So if the book designer follows the normal practice of not matching the era of type to the era of the text, where's the "twist" that Zach speaks of?

Williams Caslon is clearly 21st century, not 18th, though taking inspiration from the 18th.

I'm sorry, I thought it was a revival.

>I'm sorry, I thought it was a revival.

It is a revival, but not an attempt to look like 18th century printing. Justin Howe tried that already. My essay on I Love Typography addressed this issue.

Dolly is one of my all time favourite fonts, but I have never sen it used in any book. Does anybody have book that uses Dolly? If so, could you please scan it. I'd love to see how it looks. Dolly was designed for books, but I've only seen it in logo's and titles.

@Bill: ...not an attempt to look like 18th century printing.

Surely there is more to 18th century printing than its roughness and relative distress.

The basic letter shapes of Williams Caslon are 18th century. It looks like "Caslon", it's named Caslon, and, to a similar extent as in Adobe Caslon, it has flexed serifs, and parses the uncoated-stock press gain of 18th century printing as both a softening of the outline and a reduced-contrast "darkness" that is the theme of this thread.

The technique, a complex drawing of each glyph, suggests the variation in cutting and randomness of printing artefacts that one associates with 18th century printing, as opposed to the subsequent sharpness and regularity of the Didone style -- and is quite different than the application of a thematic, reductive system, as in the design of Matrix or Charter.

It's not a 21st century idea: Goudy was the first to develop it in Kennerley (1911).

Nick, your characterization is off the mark.

My aim in this revival was to create a look that is "contemporary classic," something with a timeless quality, rather than a design redolent of the era in which it was first created. I didn't want an "antique" or "retro" look. So my approach has been to adopt some features of the original look, which I think deserve reviving, while consciously rejecting others. Your account only notes the first, not the second.

To put it another way, my general approach to revival has been more like that of Matthew Carter's Miller, not that of Justin Howe's Founder's Caslon.

I was indeed inspired by the look of printed letter press, and have aimed to have my font, when printed by contemporary offset and direct-to-plate methods, possess some of the virtues of the letter press versions. However, I consciously rejected the 18th century look of clumpy inks and large ink spread due to printing on damp paper. My inspiration was the look of the ink and ink spread involved in early 20th century printing.

I also deliberately changed the look of 20th century metal type, as well. I did round external corners, as with letter press ink spread. However, I kept joins of branches sharp and unbracketed, contrary to the effect of letter press ink spread. Williams Caslon Text is more crisp internally, and in this feature, together with others, is a 21st century digital design, and not an 18th, 19th, or 20th century one.

The outlines, particularly of the lower case, do have a hand-modeled look. This modeling is designed to increase readability, and to achieve a welcoming, humanist aesthetic. It does not mimic the artifacts of Caslon's 18th-century punch cutting. I have seen the drawings of Mergenthaler Linotype Caslon Old Face, at the Printing Museum in Andover, which are based on blow-ups of metal type. These old Linotype drawings do indeed attempt to reproduce the irregularities of hand punch cutting. My outlines don't have this goal, and don't have these artifacts.

Bill, your Caslon is a retro design.
It is in no way a modernization or post-modernization, but firmly in the tradition of 20th century old style revivals.

How can something that it is clearly recognizably as a "Caslon" revival be anything but a retro design?!
Despite being revised at various times subsequently, the Caslon we all know and love remains an 18th century look.
"Caslon", the collective archetype, is an 18th century typeface.
Furthermore, the "modeling" you have employed, for purposes of readability, is strongly informed by the look of letterpress, which again gives the face a retro, antique look, as if the ductus, contrast, and serif shapes of Caslon were not cue enough.

@William: However, contrary to what you say, I consciously rejected the 18th century look of clumpy inks and large ink spread due to printing on damp paper.

I didn't comment on your rationale (and it should be noted that effects are not always as intended).
I said that your face "parses the uncoated-stock press gain of 18th century printing", and could have added "and earlier".
I suspect this is because you've followed the twentieth century manner of representing the appearance of old styles.
In revivals of later faces, such as didones, the tendency was generally to remove the trace of process and emphasize their simple geometry. For instance, the myth of the geometric Bodoni was long dominant, until ITC Bodoni in 1994 represented the letterpressed image.

I also, though, deliberately changed the 20th century metal look as well.

The style of rounded external corners and sharp bracket joins, which you adopted as a representation of the letterpress process, has been in use since early in the 20th century, so you didn't change anything. As I mentioned, Goudy originated it. Benton used it in Cloister, his Jenson, in 1920. Digital fonts such as Bitstream Baskerville, and Monotype Bulmer, to name two digital 18th century revivals from the late 20th century, feature it. Your Caslon is similar to Adobe Caslon in its style of finish, although the proportions are different. Filosofia went further in playing round outside corners against sharp inside corners -- your foot serifs are slightly bracketed.

I did not attempt to ape the artifacts of metal punch cutting.

Again, I made no comment on your motives. I pointed out that the irregularity, or "modeling" as you say, of your drawing (e.g. serifs are not uniformly copied and pasted, with that on the right of "n" being wider than that on the left) suggests the irregularity of punch cutting. And although your drawing does not reproduce specific gnarly artefacts, as Justin Howe's did, irregularities such as the beak of "r" that is slightly larger than the corresponding triangle in other letters (this detail is same-sized in many other Caslons) suggest an artefact of letterpress printing pressure, although that was not your intention.

Nick, there is a difference between "retro", which appeals to nostalgia, and "classic", which appeals to a sense of timeless rightness.

You don't acknowledge the reality of this category, but most of the rest of the world does.

The effort to produce better digital versions of classics has been a major movement in type, including Slimbach's efforts. And the phase of digital "new classics" with optical sizes and lower contrast is phenomenon of the past 10 or 15 years, including Bembo book and indeed various recent Caslon revivals other than mine.

When you say that Goudy already did what I did, you are simply failing to understand the issue. Goudy did letter press, and letter press just had to live with ink spread. Now with offset printing and digital type we can decide selectively what of the effects of ink spread to accept and reject. This means molding type in a different way, and new decisions on how to do it. This is only one issue involved in revivals, but it is an important one.

Your analysis is reductive and dismissive, and so in my view misses or dismisses important issues.

Nick, there is a difference between "retro", which appeals to nostalgia, and "classic", which appeals to a sense of timeless rightness. You don't acknowledge the reality of this category, but most of the rest of the world does.

I do, and I would say that any Caslon revival is both classic and retro, but I fail to see how something that is timeless can be a 21st century typeface. You can't have it both ways, Bill!

The phase of digital "new classics" with optical sizes and lower contrast is phenomenon of the past 10 or 15 years,

I disagree that it is a recent digital phenomenon, there is nothing particularly new in the process of updating the classics, it goes on all the time. The process of designing weightier serifed text faces, to compensate for the attenuation of the high-res digital-to-offset process, has been going on since the start of digital, and applies to both new and classic serifed faces. Yes, there have been more and more optically-enabled faces since 1995, but Adobe Garamond, FB Eldorado and H&FJ Didot were earlier.

When you say that Goudy already did what I did, you are simply failing to understand the issue. Goudy did letter press, and letter press just had to live with ink spread.

I have explained this before: the resolution of letterpress in Goudy's era was much higher than in the historic era (pre-didone). He was doing large drawings, not punchcutting at that time (his early career). Therefore he was able to produce letterforms which pastiched the vagaries of earlier, rougher printing. Look in Goudy's Type Designs where he described his problems with the technicians at Linotype who wanted to "clean up" his Garamond (e.g. straighten the slant on the vertical stem of "a"), because "they had always done it that way".

This means molding type in a different way, and new decisions on how to do it.

If I am being dismissive, it is not of your excellent typeface, but of claims that you are making for its originality which are exaggerated. Do you really believe that type designers for the past 100 years have not "molded" and "modeled" their drawing of type -- especially in the oldstyle genre -- to both suggest and anticipate the effect of press gain? How does your drawing differ in stylization from that of Bitstream Baskerville or Monotype Bulmer?

Here is Cloister, scanned from a 1920s magazine ad: note the heft, the molded roundedness of the serifs, and the sharpness of the internal joints.

Nick, on timelessness, I wrote that the goal is a *sense* of timeless rightness. I am well aware that because of the continual changes in technology, in order to *look* timeless, you have to keep changing. The effort to get digital revivals to actually look 'Classic' is the latest effort in a long standing one, as you say. One of the things I have done, reducing contrast, is part of the more recent efforts in revivals. I'm well aware that revivals of classics are of their time. Adobe Caslon, Founders Caslon are, and would be done differently today. I'm sure a late 19th century performance of Beethoven sounded a lot different than a late 20th century one.

I don't know how original I've claimed to be, but I have claimed to specific things differently. A lot relate to readability, which you don't believe in, so I'll write about that elsewhere.

The example you give of Cloister I'm pretty sure is big. The point about a lot of earlier digital revivals is that they were based on larger sizes, where the ink spread makes less of a difference. One of the key things about doing classics right, in my view, is to reduce contrast in the 10-12 point area, not just in the 8 point and below range, which is where ATF did it. And Slimbach also tends to have pretty high contrast in the regular range. ATF was letter press, and the 'stroke' that ink spread adds was a kind of automatic optical sizing, in my view. So where I differ with Twombly and Simbach is that I've designed a regular size with lower contrast. However, it's not that simple, because actually my joins are pretty narrow. The look is influenced by the way the weight changes through the arches.

All the things I have done on this are only subtly different, but they add up to a different look. One of the things I am proudest of is that I think you can recognize Williams Caslon on the page from its texture, without looking at details. The differences are small but systematic, so they add up to a different "look". The other thing I am pleased with is that it doesn't look like anything else, like something that came out of Reading or KABK or Adobe.

I do reject the retro label. I don't have a problem with retro, but this is not an exercise in nostalgia, and would be disappointed if that is the reaction. I hope readers will just see a comfortable, welcoming and highly readable font, not one that says "18th century" or "old fashioned." Well, ok some of the swashes may look old timey, but they're not the heart of the font, and you can use them or not for a special effect.

I do reject the retro label

You do well to do so because, strictly speaking, Retro as a style is pretty much confined to the 1950s to 1970s, a period marked by the death of letterpress and the ascendancy and triumph of the phototypesetter and offset lithography. Powdered wigs (Caslon) and snuff boxes (Scotch Roman) were not au courant at the time...

Guys, sorry to barge in here. You are both talented designers, and fascinating as your argument/discussion is, doesn't it belong on a new thread entirely?

William, you really don't have to justify or defend your design. It's a good revival for present use.. and I'm sure it's gonna look great on those books. I think Nick initially was making another point though, but this has become a back&forth off topic.

Don't spank me.

What, bringing sanity to Typophile? I am highly offended :)

Thanks fellas. Enough.

...readability, which you don't believe in...

I don't believe the readability of a typeface is measurable.

Nothing is measurable perfectly. Everything is measurable in some way (in fact, in an infinite number of ways). And none of that makes readability moot.

hhp

Just noticed all the discussion on this thread I was missing!

Nick, the twist I was referring to isn't typographic – it's in the content. When I'm close to finalizing the design of the first book, I'll probably post samples here for critique, and explain the twist then. I realize Caslon isn't a 19th century face – but as Bill said, lots of 19th century books (including many of the original editions I've been working from) were printed in Caslon.

And to weigh in on Nick and Bill's debate, with my humble two cents as a quasi-informed type consumer, what I liked about Williams Caslon was precisely the quality Bill is underlining – it's classic, and a Caslon can't be otherwise, but it does have some kind of contemporary, updated feel. I know not why, but that's my impression. It makes sense to me that Boston Magazine is using it.

...lots of 19th century books (including many of the original editions I've been working from) were printed in Caslon.

That's unlikely. In the first half of the century the style of text type was modern (didone). Mid century, some rough old Caslon type, with long "s", was used for a few historical pastiches. In the second half, a genre of modernized "old style" emerged, but it wasn't named Caslon and was a far cry from what we would recognize as Caslon today, although there are some latter day Caslons (e.g. 471) which are based on it, erroneously.

James Mosley:
http://typefoundry.blogspot.com/2009/01/recasting-caslon-old-face.html

Nick, as the Mosley article makes clear, copies of a fairly original Caslon were available in the US beginning in the 1850s. I don't know how widely they were used.

Although "Old Style" is significantly different from Caslon itself, it does indeed adapt some of Caslon's look, so that Caslon itself is not too alien to a 19th century look.

According Mac McGrew, Caslon itself became very popular in the US when Vogue magazine adopted it in 1892. Also according to him, Caslon 471 is not based on Old Style, as you say, but on Caslon itself, in the version from the electroplated version available in the US from the 1850s on. ATF acquired this as part of the consolidation of foundries at the end of the 19th century.

I don't know how much 471 and the later 540, with reduced descenders, were recut from the electroplated version, but I am guessing that ATF redid it and regularized it with the Pantographic punch cutter. The digital 540 is based on the larger sizes of Caslon, not the smaller sizes, which is why it works well as a display face, but not a text face.

@Bill: Caslon 471 is not based on Old Style,

"Caslon 471 was designed by the staff of American Type Founders as their first revival of Caslon. It is based on the Old Style No. 1 typeface used in an 1865 specimen book from the L.J. Johnson foundry in Philadelphia."

That's from the Caslon wiki. Not always a reliable source. However, unless one were to inspect these types closely, I suspect it would be hard to say just how much "Caslon" of what size there is in them.

Anyway Bill, although there was some marginal use of more authentic Caslon types in the 19th century (and especially towards the end), I was suggesting to Zach that what he is seeing as book-text "Caslon" in the original editions of lots of 19th century books is probably a modernized old face.

Although "Old Style" is significantly different from Caslon itself, it does indeed adapt some of Caslon's look, so that Caslon itself is not too alien to a 19th century look.

That's pretty sloppy, given the variation in Caslon's types between sizes. This argument appears to be that all old styles that aren't Garamond, Jenson or Bembo are Caslon! As an alternative to the usual Shavian eulogy, how about a little Homer (Simpson, that is): "Caslon, is there anything it can't do?"

I would say that Caslon itself was indeed alien to the 19th century, because it was revived in the 1840s for historical pastiche and found wanting for any other use.

Nick, the Old Style from Johnson was indeed the electroplated Caslon. Mac McGrew explains that while it was Caslon, "Johnson simply called the face Old Style, as family names were a later development." But it is, in fact Caslon, and not a font derived from the Miller and Richard Old Style Antique designed by Alexander Phemister in around 1860.

And hence, Caslon 471 is indeed based on the originals, but I think recut, and not the same.

By the way, that also shows you that the matrices from the Electroplated Caslon were around in the US since the 1860s. As I said, how often they were used, I don't know.

The real Caslon craze it seems only started in the 1890s, according McGrew.

We've disagreed before about to what extent the Phemister Old Style face resembles Caslon—where I'm also not in full agreement with Mosley——so I'll just let that lay.

Nick, you may well be right – the (digitized) originals I'm working from have something that looks Caslon-ish, but it's hard for my eyes to tell, due to a combination of lack of experience, laziness, and poor quality scanning. Here, for example, is a paragraph from Google Books' 1897 copy of "Dracula":

I look at that and think "Hard to tell, looks like Caslon, or something in the same ballpark at any rate" :)

Here's that paragraph in Bill's Calson for comparison:

(Though as a side note, a week ago I decided I'd use Storm Walbaum for Dracula, and probably Williams Caslon for some of the other books. Unless Nick keeps giving me second thoughts...)

It seems like the "Old Style" the article refers to is the same face that I've seen in a lot of early 20th century books, but never known the name of, e.g.:

Zach, I do think your samples are "Old Style", meaning one of the faces derived from the 1860 Old Style Antique of Alexander Phemister.

As the Jaspert Encyclopedia of Typefaces explains, "Its purpose was to cut a type that overcame what was regarded as archaic in CASLON. The serifs are slighter and more sharply cut. Stress is vertical, though not abruptly so. ...Ascenders and descenders are rather short."

In other words, Old Style was the first Caslon revival, though it was undoubtedly strongly influenced by current 19th century fashions, as Mosley points out. As you I think rightly note, it also does distinctly remind one of Caslon, even though it is substantially changed in some respects.

As I wrote earlier, that a re-publication of a text has to be in the same type as the era in which it was written is I think a questionable notion—and in any case not a generally accepted one. I can understand not wanting it to look totally anachronistic, but aside from that, I think the main thing is to be readable, look good, and suit the mood of the piece.

My problem with Old Style and the many fonts derived from it is that I think they look dull. If you test one of the digital ones—there are a bunch—I think you will see what I mean. For me the original contrast is just a little too low, and the ascenders too short—which, as Tracy notes, tends to make a face dull. Also short ascenders looks less "literary" in a book setting. I personally think they are less readable, but I know my view is very debatable.

Just looking now at some of the digital versions, they seem rather high contrast compared to 19th century printed versions. See Bruce Old Style, Binny Old Style, and Old Style 7. Bookman is closer in being low contrast, but I think rarely used for books. Century Old Style is end 19th century and not so closely related to Caslon, but has perhaps more the feel of that era, if that's what you want.

ps. I would use old style figures with text. I notice that the "Old Style" font has them...

> As I wrote earlier, that a re-publication of a text has to be in the same type as the era in which it was written is I think a questionable notion—and in any case not a generally accepted one. I can understand not wanting it to look totally anachronistic, but aside from that, I think the main thing is to be readable, look good, and suit the mood of the piece.

I agree, though looking anachronistic is a somewhat subjective experience. On Dracula I ended up switching to Walbaum because -- bringing us back to the original thread topic -- it had a darker look, both in color and tone, which I think fits well with the text, even though it wasn't written in Germany or in Walbaum's era.

...that a re-publication of a text has to be in the same type as the era in which it was written is I think a questionable notion—and in any case not a generally accepted one.

This is certainly true for those who don't know any better.
As long as a type looks suitably old to the reader it will do.
However, for those who are aware of the vintage of both text and type, there is a dialectic present, which one would hope has been considered by the typographer, because it effects the meaning of the work.

I wouldn't say that type has to match the era of the text, but if it doesn't, there should be a significant reason, not just "close enough".

And for those who do know better, the idea of "matching" is simplistic.

If you exactly mimic the look of an earlier era it will generally look antiquarian. If that's what you want, fine, but it won't carry the same meaning as it did in the era in which it was originally printed, for the printing then looked "normal." In other words, true "matching" is, ironically, not authentic to the original feel, but adds a message of antiquity or nostalgia.

So the "dialectic" you rightly speak of should start with a recognition that mimicking is as a rule a poor idea. It is mimicking that requires a special reason, and not the use of a type that today looks "normal." And if you choose not to mimic, then a wide range of possibilities opens up, many of which may be good ones.

Historical association is one factor, but I think it is generally best attenuated. If I remember rightly the Penguin Complete Shakespeare is in Plantin. But it isn't the original look of the first printing of Shakespeare, but of the 20th century revival, which looks "classic" and 20th century, when it was published. So the historical association is rightly attenuated by quite a bit.

Matching is not mimicking, which is besides impossible unless one prints letterpress.

I'm talking about historically sensitive type choices that express something of the aesthetic of an age, apart from considerations of newness or antiquity.

What is the relationship between the shape of a sentence and the shape of a type?

I would like to believe there is some kind of ecological synergy, that writers have some inkling of the appearance of their text in the publications of their day, and write accordingly. Cultural works do not exist in isolation from time and place, but are expressions of them. If today we are interested in the past at all, and want to feel and understand it, surely inserting a filter of anachronism between text and reader is no help.

This is why historical theatre and film is generally performed in period costume. Typographers should pay attention to that, wise up, and stop being so sloppy and self-indulgent.

Nick, your analogy between costume in a play or movie and a typeface is interesting, but also I think brings out the differences. Costumes don't "disappear" as you see the play or movie to the extent that text typefaces do as you read the book.

I believe Barnes and Noble has published a bunch of paperback "Classics", popular out-of-copyright books, and I will have to check, but I believe they have rather similar formats and fonts. I really don't think that is sloppy and self-indulgent. It may be that they put a priority on the look of the overall series, and frankly I don't think it would interfere with the reader's enjoyment, provided they are well set otherwise.

I'm not saying that historical allusion is bad, but I just think you are overdoing its importance, when there are other priorities.

Costumes don't "disappear" as you see the play or movie to the extent that text typefaces do as you read the book.

Sure they do.
In pursuit of the suspension of disbelief, set designer, costume designer, makeup artist, hair stylist and art director work under the director to create a seamless whole that enables the audience to become engrossed in the story.

I wouldn't say I over-represent the importance of historical allusion, given that it is generally so poorly adhered to.

Nick, let me try again on costumes vs type.

I don't think reading the novel or play in Times New Roman vs Baskerville is going to make you think that differently about the characters in Huckleberry Finn or Hamlet. But a great film that has costumes and the look of an actor may make you think about these stories differently. I just think the influence of text type is less specific, less attached to the content of the story. It sets a mood rather than gives a specific historical reference. It is a step further removed from the content of the experience.

I think the mood is more important than an historical connection. So, for example, I like Zach's idea of using Walbaum for Dracula, because it is a little unusual looking and stylish, so it might add to the general creepiness of the story. That Dracula was, in 1897, originally set in an Old Style (I checked on Google Books) I don't think is as important as the mood.

The historical reference you can do more with display type, and I'm not against it. I agree with you that historical setting should be considered, but I have the feeling that I would put it lower in priority and you do, and I also think there's a danger of looking antiquarian, which will may work against the mood, in spite of being historically more accurate.

Speaking of Dracula, Murnau's use of text (not just in the intertitles) in Nosferatu was brilliant.

Don't know about Walbaum.
I would prefer to read Dracula in the original Old Style, antiquarian that I am.